Full release in Q3; supports containerisation, machine learning
Data management specialists Hortonworks today announced version 3.0 of the company’s Apache Hadoop-based Hortonworks Data Platform (HDP).
HDP 3.0, which helps users store, process and analyse large volumes of data in a wide range of formats,was announced at the Dataworks Summit in San Jose.
It is based on the latest version of open source distributed data storage and processing toolkit Apache Hadoop 3.1, and now offers containerisation, along with support for deep learning workloads like TensorFlow.
Full release is slated for Q3.
Hortonworks said that the platform includes engineered support for all of the major cloud object stores: Amazon S3 with support for native EDW, Windows Azure Storage Blob (WASB), and Google Cloud Storage (GCS).
HDP 3.0 – enabled via Apache Hive 3.0 – offers the only unified SQL solution that can perform interactive queries at scale, regardless of whether the data lives on-premises or in the cloud, Hortonworks said.
The release of HDP 3.0 seeks to meet rising developer demand for the ability to build and deploy data-intensive apps at pace, as well as support deep learning apps and workloads including machine learning platforms TensorFlow and Caffe, among a range of other features. It now supports third-party applications run in Docker containers, as well as native YARN containers, the company added.
It also features erasure coding that boosts storage efficiency by 50 percent and a real-time database that eliminates the performance gap between low latency and high throughput workloads to process more data at a faster rate. A big focus of the release is meeting the needs of hybrid cloud users, and the release includes cloud storage support to store endless amounts of data in its native format.
HDP 3.0’s real-time database query optimisations are underpinned by “best-of breed” Micron NVMe Solid State Drives, Eric Endebrock, VP Storage Solutions Marketing at Micron Technology added.
Trio of New Agreements
The release comes as the company also announced that it had extended its relationships with Microsoft Azure and Google Cloud.
It has also announced a new hosted data management and analytics offering; made available as a service via IBM Cloud.
(IBM Hosted Analytics with Hortonworks (IHAH) is a fully-provisioned data management and analytic environment via IBM Cloud, comprising HDP, IBM Db2 Big SQL, and the IBM Data Science Experience.)